Discussion:
repetitive octane 2 hangs
(too old to reply)
m***@ju.edu.jo
2006-07-26 10:45:29 UTC
Permalink
I have an Octane 2 with a 600 MHz MIPS R14000 processor (1.0 GB RAM)
running Irix 6.5.22m. It is in an air conditioned room (22-28 degrees
C).
It was recently cleaned with compressed air, however, only without
openning the case. It is comnnected to a UPS to exclude possible
electrical instabilities. Last week it started to repetitively hang,
it hanged nearly once per day, especially at night. When I removed
virtual swaps, it remained OK over more than 48 hours, then it hanged
again. Once it hangs the only way to restart it is to remove its power
supply. The SYSLOG contains no odd message at the time of the
freeze. Upon running the hardware diagnostics, all hardware
parts (cpu, memory, etc...) pass the tests.
I tried to measure the time it takes from one hang to another and
I found it takes around 56 hours, it looks to be very precise in its
run-hang cycles.
If we try to ping or telnet or ftp the machine during its hang the
order simply fails. Anybody have any idea about this? please help me
it is really very urgent!.
Gary Heston
2006-07-27 01:23:29 UTC
Permalink
Post by m***@ju.edu.jo
I have an Octane 2 with a 600 MHz MIPS R14000 processor (1.0 GB RAM)
running Irix 6.5.22m. It is in an air conditioned room (22-28 degrees
C).
It was recently cleaned with compressed air, however, only without
openning the case. [ ... ]
Now, open it up and find out what you blew all the loose lint dust
into (probably something to do with the CPU cooling). Blow out the
power supply via every opening it has, make sure all the heatsinks
are clean as well. You'll want to do this outdoors, and follow ESD
protection practices (antistatic mat, wrist strap, etc.).

You can't properly clean a system without opening it up.


Gary
--
Gary Heston ***@hiwaay.net What do you call two SUVs colliding?
Poetic justice.
A worthwhile endeavour:
http://www.thebreastcancersite.com/cgi-bin/WebObjects/CTDSites
Josef Scheele
2006-07-28 11:12:44 UTC
Permalink
Post by m***@ju.edu.jo
I tried to measure the time it takes from one hang to another and
I found it takes around 56 hours, it looks to be very precise in its
run-hang cycles.
A cron job that kills the kernel or daemon ?
m***@ju.edu.jo
2006-07-29 14:18:52 UTC
Permalink
Post by Josef Scheele
Post by m***@ju.edu.jo
I tried to measure the time it takes from one hang to another and
I found it takes around 56 hours, it looks to be very precise in its
run-hang cycles.
A cron job that kills the kernel or daemon ?
When I performed file reorganization job (fsr_XFS), the machine stopped
hanging and its OK now.., do you have any idea about this!?

Loading...